Update to fms@2023.02.01 #1874

climbfuji · 2023-08-18T03:02:30Z

Description

The spack-stack dev team is hoping to include the update to the newly released fms@2023.02 in the next spack-stack release (early September). I am wondering if the ufs-weather-model is able to move up.

I built fms@2023.02 in spack-stack with the legacy fms_io (i.e. version 1) interface enabled (it's now off by default) on my macOS, and I was able to compile the ufs-weather-model develop as of two days ago with that version loaded. I haven't run any tests though.

Solution

Please let us know if a move up to fms@2023.02 at some point in September is possible. The ufs-weather-model wouldn't have to move to that version immediately. Only when updating from the spack-stack-1.4.1 modules (plus PIO updates in the chained environment) to 1.5.0 the new FMS version would be used.

Alternatives

We can also install the existing fms version side by side with the new one to make the process of updating fms in ufs-w-m even easier.

Related to

See JCSDA/spack-stack#709

The text was updated successfully, but these errors were encountered:

junwang-noaa · 2023-08-18T12:07:59Z

@climbfuji Thanks for building fms 2023.02 in spack-stack. We hope we can switch to fms2 io when using fms 2023.02 without using -Duse_deprecated_io option. Some updates are required in several components, you can see the NeedsPort.txt list and discussion in the fv3 dycore PR#272

climbfuji · 2023-08-18T12:53:22Z

JEDI will need -Duse_deprecated_io for some time, thus we have to build it with that option turned on.

junwang-noaa · 2023-08-28T15:18:08Z

@Hang-Lei-NOAA would you please make request to install FMS 2023.02 on wcoss2?

junwang-noaa · 2023-09-11T13:55:09Z

@climbfuji @natalie-perlin Is there an FMS 2023.02 test version available on hera for people to test it in UFS? Thanks

climbfuji · 2023-09-11T13:56:06Z

Yes it will be available as part of spack-stack-1.5.0.

junwang-noaa · 2023-09-11T14:05:22Z

@climbfuji Have you tested it in UFS WM? Is it working? I am asking because we need to test it first in order to make request to NCO to install it on wcoss2, which may take some time.

climbfuji · 2023-09-11T14:16:40Z

Once, but it needs to be re-tested (by someone else than me). spack-stack-1.5.0 by default uses fms@2023.01, but 2023.02 is available as an alternative.

Hang-Lei-NOAA · 2023-09-11T14:30:17Z

@jun Wang - NOAA Federal ***@***.***> I have prepared to send the 2023.02 to NCO. Since our request to install spack-stack will be delayed.

…

On Mon, Sep 11, 2023 at 10:16 AM Dom Heinzeller ***@***.***> wrote: Once, but it needs to be re-tested (by someone else than me). spack-stack-1.5.0 by default uses ***@***.***, but 2023.02 is available as an alternative. — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFC7JOOELJZF4CCHIZLXZ4MNFANCNFSM6AAAAAA3U5IYJ4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

Hang-Lei-NOAA · 2023-10-04T15:11:38Z

Dear all, NCO has installed the fms/2023.02 on wcoss2 cactus and dogwoods for testing and further will move them to NCO operational structure. Please test at module use /apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9 module load fms/2023.02 On Mon, Sep 11, 2023 at 10:30 AM Hang Lei - NOAA Affiliate < ***@***.***> wrote:

…

@jun Wang - NOAA Federal ***@***.***> I have prepared to send the 2023.02 to NCO. Since our request to install spack-stack will be delayed. On Mon, Sep 11, 2023 at 10:16 AM Dom Heinzeller ***@***.***> wrote: > Once, but it needs to be re-tested (by someone else than me). > spack-stack-1.5.0 by default uses ***@***.***, but 2023.02 is available > as an alternative. > > — > Reply to this email directly, view it on GitHub > <#1874 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AKWSMFC7JOOELJZF4CCHIZLXZ4MNFANCNFSM6AAAAAA3U5IYJ4> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

junwang-noaa · 2023-10-13T14:29:07Z

@Hang-Lei-NOAA I tried to compile the fms 2023.02, I got following error:

Due to MODULEPATH changes, the following have been reloaded:
  1) netcdf/4.7.4

The following have been reloaded with a version change:
  1) cray-mpich/8.1.9 => cray-mpich/8.1.12     3) hdf5/1.10.6 => hdf5/1.14.0
  2) craype/2.7.8 => craype/2.7.13


Currently Loaded Modules:
  1) craype-x86-rome     (H)   7) nco/4.7.9           13) fms/2023.02
  2) libfabric/1.11.0.0. (H)   8) PrgEnv-intel/8.1.0  14) scotch/7.0.3
  3) craype-network-ofi  (H)   9) intel/19.1.3.304    15) ufs_wcoss2.intel
  4) envvar/1.0               10) craype/2.7.13       16) hdf5/1.14.0
  5) cray-pals/1.2.2          11) cray-mpich/8.1.12   17) netcdf/4.7.4
  6) python/3.8.6             12) cmake/3.20.2

As you can see other libraries such as ESMF/MAPL... are not loaded. I think the fms library should be built with cray-mpich/8.1.12 with netcdf 4.9.2. My module file on dogwood at:

/lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231013/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua

Thanks

Hang-Lei-NOAA · 2023-10-13T14:51:41Z

@jun Wang - NOAA Federal ***@***.***> Do you remember that NCO was installing the netcdf and hdf5 etc that we are using into: module load netcdf-A/4.9.2 module load hdf5-A/1.14.0 This module is built with that one. And there is no preload request in NCO's modulefile. /apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9/fms/2023.02.lua

…

On Fri, Oct 13, 2023 at 10:29 AM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> I tried to compile the fms 2023.02, I got following error: Due to MODULEPATH changes, the following have been reloaded: 1) netcdf/4.7.4 The following have been reloaded with a version change: 1) cray-mpich/8.1.9 => cray-mpich/8.1.12 3) hdf5/1.10.6 => hdf5/1.14.0 2) craype/2.7.8 => craype/2.7.13 Currently Loaded Modules: 1) craype-x86-rome (H) 7) nco/4.7.9 13) fms/2023.02 2) libfabric/1.11.0.0. (H) 8) PrgEnv-intel/8.1.0 14) scotch/7.0.3 3) craype-network-ofi (H) 9) intel/19.1.3.304 15) ufs_wcoss2.intel 4) envvar/1.0 10) craype/2.7.13 16) hdf5/1.14.0 5) cray-pals/1.2.2 11) cray-mpich/8.1.12 17) netcdf/4.7.4 6) python/3.8.6 12) cmake/3.20.2 As you can see other libraries such as ESMF/MAPL... are not loaded. I think the fms library should be built with cray-mpich/8.1.12 with netcdf 4.9.2. My module file on dogwood at: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231013/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua Thanks — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFEHUKIJKXI3X7BHFSLX7FF35AVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRRGYYTQMBUGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

Hang-Lei-NOAA · 2023-10-13T14:59:10Z

For all, there is a misunderstanding to be clarified for users. NCO never used hpc-stack on the operational machines. We just used a special version of hpc-stack to deliver the code and facilitate the installation process. NCO built the libraries under hpc-stack and move them into NCO vertical structures. We are actually using NCO vertical structures on operational machines. This is the same result as that we deliver code to NCO through original tarballs. e.g. some tarball deliveries are also in NCO vertical structure. scotch/7.0.3; ecbuild; gftl_shared etc. NCO use their own modulefiles. On Fri, Oct 13, 2023 at 10:51 AM Hang Lei - NOAA Affiliate < ***@***.***> wrote:

…

@jun Wang - NOAA Federal ***@***.***> Do you remember that NCO was installing the netcdf and hdf5 etc that we are using into: module load netcdf-A/4.9.2 module load hdf5-A/1.14.0 This module is built with that one. And there is no preload request in NCO's modulefile. /apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9/fms/2023.02.lua On Fri, Oct 13, 2023 at 10:29 AM Jun Wang ***@***.***> wrote: > @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> I tried to compile the > fms 2023.02, I got following error: > > Due to MODULEPATH changes, the following have been reloaded: > 1) netcdf/4.7.4 > > The following have been reloaded with a version change: > 1) cray-mpich/8.1.9 => cray-mpich/8.1.12 3) hdf5/1.10.6 => hdf5/1.14.0 > 2) craype/2.7.8 => craype/2.7.13 > > > Currently Loaded Modules: > 1) craype-x86-rome (H) 7) nco/4.7.9 13) fms/2023.02 > 2) libfabric/1.11.0.0. (H) 8) PrgEnv-intel/8.1.0 14) scotch/7.0.3 > 3) craype-network-ofi (H) 9) intel/19.1.3.304 15) ufs_wcoss2.intel > 4) envvar/1.0 10) craype/2.7.13 16) hdf5/1.14.0 > 5) cray-pals/1.2.2 11) cray-mpich/8.1.12 17) netcdf/4.7.4 > 6) python/3.8.6 12) cmake/3.20.2 > > As you can see other libraries such as ESMF/MAPL... are not loaded. I > think the fms library should be built with cray-mpich/8.1.12 with netcdf > 4.9.2. My module file on dogwood at: > > > /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231013/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua > > Thanks > > — > Reply to this email directly, view it on GitHub > <#1874 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AKWSMFEHUKIJKXI3X7BHFSLX7FF35AVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRRGYYTQMBUGQ> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

junwang-noaa · 2023-10-13T15:00:47Z

@Hang-Lei-NOAA can you let me know how to load hdf5-A and netcdf-A? My module file is at:

/lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231013/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua

Thanks

Hang-Lei-NOAA · 2023-10-13T15:18:52Z

@jun Wang - NOAA Federal ***@***.***> I edit it into: /lfs/h2/emc/eib/save/Hang.Lei/ufs_wcoss2.intel.lua

…

On Fri, Oct 13, 2023 at 11:01 AM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> can you let me know how to load hdf5-A and netcdf-A? My module file is at: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231013/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua Thanks — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFBQLUZUHDZEI3X5A4DX7FJSVAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONRRGY3DKOBTGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-10-13T18:36:32Z

@Hang-Lei-NOAA These hdf-A/netcdf-A libraries are not loaded. Here is what I got:

Due to MODULEPATH changes, the following have been reloaded:
  1) hdf5/1.10.6     2) netcdf/4.7.4

The following have been reloaded with a version change:
  1) cray-mpich/8.1.9 => cray-mpich/8.1.12     2) craype/2.7.8 => craype/2.7.13


Currently Loaded Modules:
  1) craype-x86-rome     (H)   7) nco/4.7.9           13) fms/2023.02
  2) libfabric/1.11.0.0. (H)   8) PrgEnv-intel/8.1.0  14) scotch/7.0.3
  3) craype-network-ofi  (H)   9) intel/19.1.3.304    15) ufs_wcoss2.intel
  4) envvar/1.0               10) craype/2.7.13       16) hdf5/1.10.6
  5) cray-pals/1.2.2          11) cray-mpich/8.1.12   17) netcdf/4.7.4
  6) python/3.8.6             12) cmake/3.20.2
...
CMake Error at /apps/spack/cmake/3.20.2/intel/19.1.3.304/utnbptm3hrf7gppztidueu4jogfgemut/share/cmake-3.20/Modules/FindPackageHandleStandardArgs.cmake:230 (message):
  Could NOT find ESMF (missing: ESMF_LIBRARY_LOCATION
  ESMF_INTERFACE_LINK_LIBRARIES ESMF_F90COMPILEPATHS) (Required is at least
  version "8.3.0")
Call Stack (most recent call first):
  /apps/spack/cmake/3.20.2/intel/19.1.3.304/utnbptm3hrf7gppztidueu4jogfgemut/share/cmake-3.20/Modules/FindPackageHandleStandardArgs.cmake:594 (_FPHSA_FAILURE_MESSAGE)
  CMakeModules/Modules/FindESMF.cmake:121 (find_package_handle_standard_args)
  CMakeLists.txt:144 (find_package)

The error file is at:
/lfs/h2/emc/ptmp/jun.wang/FV3_RT/rt_102039/com*l/err

junwang-noaa · 2023-10-23T14:24:12Z

@climbfuji Can we change the issue to "update to fms@2023.02.01" or if you want me to create a separate issue for "updating to fms@2023.02.01? Thanks

climbfuji · 2023-10-23T14:36:46Z

@climbfuji Can we change the issue to "update to fms@2023.02.01" or if you want me to create a separate issue for "updating to fms@2023.02.01? Thanks

@junwang-noaa I changed it.

Hang-Lei-NOAA · 2023-10-24T18:38:20Z

@jun Wang - NOAA Federal ***@***.***> NCO has installed the fms/2023.02.01 on para, Please test it on dogwoods.Then they will move it to prod /apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9/fms/2023.02.01.lua module use /apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9 module load fms/2023.02.01 ~> ls /apps/ops/para/libs/intel/19.1.3.304/cray-mpich/8.1.9/fms/2023.02.01 include_r4 include_r8 lib

…

On Mon, Oct 23, 2023 at 10:37 AM Dom Heinzeller ***@***.***> wrote: @climbfuji <https://github.com/climbfuji> Can we change the issue to "update to ***@***.***" or if you want me to create a separate issue for "updating to ***@***.***? Thanks @junwang-noaa <https://github.com/junwang-noaa> I changed it. — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFEBVGJ5EX3JZRWZR2DYAZ6ITAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZVGM2TCMBRGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-10-25T02:03:11Z

@Hang-Lei-NOAA The compile failed with error:

/lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc
essible.   [OPEN_NAMELIST_FILE]
      use fms_mod,           only: check_nml_error, open_namelist_file, close_file
----------------------------------------------------^
/lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc
essible.   [CLOSE_FILE]
      use fms_mod,           only: check_nml_error, open_namelist_file, close_file
------------------------------------------------------------------------^
compilation aborted for /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90 (code 1)

Did you compile the library with "-Duse_deprecated_io" option? Please see @climbfuji 's comment above.

> JEDI will need `-Duse_deprecated_io` for some time, thus we have to build it with that option turned on.

I'd suggest you check the fms/2023.02.01 compile options used in spack-stack. Thanks

Hang-Lei-NOAA · 2023-10-25T03:19:40Z

@jun Wang - NOAA Federal ***@***.***> I will coordinate with NCO to update it with the option. Thanks

…

On Tue, Oct 24, 2023 at 10:03 PM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> The compile failed with error: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc essible. [OPEN_NAMELIST_FILE] use fms_mod, only: check_nml_error, open_namelist_file, close_file ----------------------------------------------------^ /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc essible. [CLOSE_FILE] use fms_mod, only: check_nml_error, open_namelist_file, close_file ------------------------------------------------------------------------^ compilation aborted for /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90 (code 1) Did you compile the library with "-Duse_deprecated_io" option? Please see @climbfuji <https://github.com/climbfuji> 's comment above. > JEDI will need `-Duse_deprecated_io` for some time, thus we have to build it with that option turned on. I'd suggest you check the fms/2023.02.01 compile options used in spack-stack. Thanks — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFGOGXP2PUKBICLUEVTYBBXOZAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZYGM3DANRSGM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

Hang-Lei-NOAA · 2023-10-25T16:46:08Z

@jun Wang - NOAA Federal ***@***.***> NCO has updated the installation of fms/2023.02.01 Please test it as soon as possible. Thanks. On Tue, Oct 24, 2023 at 11:19 PM Hang Lei - NOAA Affiliate < ***@***.***> wrote:

…

@jun Wang - NOAA Federal ***@***.***> I will coordinate with NCO to update it with the option. Thanks On Tue, Oct 24, 2023 at 10:03 PM Jun Wang ***@***.***> wrote: > @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> The compile failed > with error: > > /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc > essible. [OPEN_NAMELIST_FILE] > use fms_mod, only: check_nml_error, open_namelist_file, close_file > ----------------------------------------------------^ > /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90(45): error #6580: Name in only-list does not exist or is not acc > essible. [CLOSE_FILE] > use fms_mod, only: check_nml_error, open_namelist_file, close_file > ------------------------------------------------------------------------^ > compilation aborted for /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/FV3/atmos_cubed_sphere/model/multi_gases.F90 (code 1) > > > Did you compile the library with "-Duse_deprecated_io" option? Please see > @climbfuji <https://github.com/climbfuji> 's comment above. > > > JEDI will need `-Duse_deprecated_io` for some time, thus we have to build it with that option turned on. > > I'd suggest you check the fms/2023.02.01 compile options used in > spack-stack. Thanks > > — > Reply to this email directly, view it on GitHub > <#1874 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AKWSMFGOGXP2PUKBICLUEVTYBBXOZAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZYGM3DANRSGM> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

junwang-noaa · 2023-10-26T00:47:26Z

@Hang-Lei-NOAA Now the model compiled successfully, but the results show that issue is not fixed. Would you please check if it is fms 2023.02.01 installed, not fms/2023.02? It looks to me it's fms 2023.02. We need the patch in fms/2023.02.01 to fix the problem. Thanks

Hang-Lei-NOAA · 2023-10-26T13:58:41Z

@jun Wang - NOAA Federal ***@***.***> Could you please test my build on dogwoods before we further check with NCO? Please also "module show fms/2023.02.01" to make sure that you load the current lib. module use /lfs/h2/emc/eib/save/Hang.Lei/forgdit/install/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9 module load fms/2023.02.01

…

On Wed, Oct 25, 2023 at 8:47 PM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> Now the model compiled successfully, but the results show that issue is not fixed. Would you please check if it is fms 2023.02.01 installed, not fms/2023.02? It looks to me it's fms 2023.02. We need the patch in fms/2023.02.01 to fix the problem. Thanks — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFHYODLN4KAWNSHBGDLYBGXKVAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBQGI2DQNBZGI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

Hang-Lei-NOAA · 2023-10-26T17:03:30Z

@jun Wang - NOAA Federal ***@***.***> NCO also update their build. You can directly test the NCO installation. Thanks On Thu, Oct 26, 2023 at 9:58 AM Hang Lei - NOAA Affiliate ***@***.***> wrote:

…

@jun Wang - NOAA Federal ***@***.***> Could you please test my build on dogwoods before we further check with NCO? Please also "module show fms/2023.02.01" to make sure that you load the current lib. module use /lfs/h2/emc/eib/save/Hang.Lei/forgdit/install/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9 module load fms/2023.02.01 On Wed, Oct 25, 2023 at 8:47 PM Jun Wang ***@***.***> wrote: > @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> Now the model compiled > successfully, but the results show that issue is not fixed. Would you > please check if it is fms 2023.02.01 installed, not fms/2023.02? It looks > to me it's fms 2023.02. We need the patch in fms/2023.02.01 to fix the > problem. Thanks > > — > Reply to this email directly, view it on GitHub > <#1874 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AKWSMFHYODLN4KAWNSHBGDLYBGXKVAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBQGI2DQNBZGI> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

junwang-noaa · 2023-10-26T17:15:46Z

@Hang-Lei-NOAA I just tested your installation, the results look good. I am testing the NCO's installation now.

junwang-noaa · 2023-10-26T18:21:15Z

@Hang-Lei-NOAA The NCO installation still does not work. So I have this in the module file:

+prepend_path("MODULEPATH", "/apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9")
+fms_ver=os.getenv("fms_ver") or "2023.02.01"
+load(pathJoin("fms",fms))
+--prepend_path("MODULEPATH", "/lfs/h2/emc/eib/save/Hang.Lei/forgdit/install/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9")
+--fms_ver=os.getenv("fms_ver") or "2023.02.01"
+--load(pathJoin("fms",fms))

The top one in /apps/ops/para does not work properly, but the bottom one under /lfs/h2/emc/eib/save/Hang.Lei works.

Hang-Lei-NOAA · 2023-10-27T13:41:18Z

@jun Wang - NOAA Federal ***@***.***> Could you please point me to your test directory? I carefully compared the NCO build log and my build log yesterday and today. They are the same in settings. I have to figure out what NCO made wrong on their end. I hope that it is not something like the previous impact of default root loading.

…

On Thu, Oct 26, 2023 at 2:21 PM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> The NCO installation still does not work. So I have this in the module file: +prepend_path("MODULEPATH", "/apps/ops/para/libs/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9") +fms_ver=os.getenv("fms_ver") or "2023.02.01" +load(pathJoin("fms",fms)) +--prepend_path("MODULEPATH", "/lfs/h2/emc/eib/save/Hang.Lei/forgdit/install/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.9") +--fms_ver=os.getenv("fms_ver") or "2023.02.01" +--load(pathJoin("fms",fms)) The top one in /apps/ops/para does not work properly, but the bottom one under /lfs/h2/emc/eib/save/Hang.Lei works. — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFGMWSQJPQLVAKYYPTLYBKS2LAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBRGYZDIMJTG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-10-27T13:55:54Z

@Hang-Lei-NOAA My code is on dogwood at:

/lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model

module file: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua

You can see the compile log from your installation at:
/lfs/h2/emc/ptmp/jun.wang/FV3_RT/rt_206918/compile_s2swa_32bit_pdlib_intel/out

from NCO's installation is at:
/lfs/h2/emc/ptmp/jun.wang/FV3_RT/rt_88930/compile_s2swa_32bit_pdlib_intel/out

It looks to me fms/2023.01 is loaded in NCO's installation even though I specified fms 2023.02.01 in the module file. While in your installation, fms/2023.02.01 is loaded. Thanks

Hang-Lei-NOAA · 2023-11-01T19:23:41Z

@jun Wang - NOAA Federal ***@***.***> It is updated. Please resubmit the job. Thanks

…

On Fri, Oct 27, 2023 at 9:56 AM Jun Wang ***@***.***> wrote: @Hang-Lei-NOAA <https://github.com/Hang-Lei-NOAA> My code is on dogwood at: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model module file: /lfs/h2/emc/nems/noscrub/jun.wang/ufs-weather/20231024/ufs-weather-model/modulefiles/ufs_wcoss2.intel.lua You can see the compile log from your installation at: /lfs/h2/emc/ptmp/jun.wang/FV3_RT/rt_206918/compile_s2swa_32bit_pdlib_intel/out from NCO's installation is at: /lfs/h2/emc/ptmp/jun.wang/FV3_RT/rt_88930/compile_s2swa_32bit_pdlib_intel/out It looks to me fms/2023.01 is loaded in NCO's installation even though I specified fms 2023.02.01 in the module file. While in your installation, fms/2023.02.01 is loaded. Thanks — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFBLIY3EGAUEV5WIJJTYBO4PLAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOBSHE3DENZZHA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-11-02T00:01:47Z

@Hang-Lei-NOAA The test runs successfully, the results look good now.

junwang-noaa · 2023-11-06T14:12:11Z

@Hang-Lei-NOAA is fms/2023.02.01 installation on wcoss2 finished yet? Would you please let me know the final module directory? Thanks

junwang-noaa · 2023-11-06T14:15:19Z

@climbfuji Is it decided that fms 2023.03 will be installed in in spack-stack 1.5.1? Does anyone test the cpld mixed mode test and confirmed the results yet? Thanks

climbfuji · 2023-11-06T14:27:05Z

@climbfuji Is it decided that fms 2023.03 will be installed in in spack-stack 1.5.1? Does anyone test the cpld mixed mode test and confirmed the results yet? Thanks

So far it's only available on Hercules, in addition to 2023.02.1 (which is the default in spack-stack-1.5.1 and available on all platforms):

[role-epic@Hercules:/work/noaa/epic/role-epic/spack-stack/orion/spack-stack-1.5.1/envs/unified-env]# find ./ -type d -name 'fms-*'
./install/intel/2022.0.2/fms-2023.02.01-bagqfo6
./install/intel/2022.0.2/fms-release-jcsda-np4324l
./install/intel/2022.0.2/fms-2023.01-7qgjvui
./install/gcc/10.2.0/fms-2023.01-rxbif7m
./install/gcc/10.2.0/fms-release-jcsda-hloh2uu
./install/gcc/10.2.0/fms-2023.02.01-2eesnun

We have an issue open for adding 2023.03 on the other platforms: JCSDA/spack-stack#860

As soon as someone from the UFS does the acceptance testing on Hercules and tells us that this version 2023.03 is ok, we will start adding it on the other platforms.

junwang-noaa · 2023-11-06T14:58:50Z

Thank you, I will create an issue for testing fms 2023.03 in UFS WM.

Hang-Lei-NOAA · 2023-11-06T15:01:32Z

@jun Wang - NOAA Federal ***@***.***> It is already in the para path that I previously provide to you on both machines. To be on prod path, it has to be waiting for the next system maintianence

…

On Mon, Nov 6, 2023 at 9:59 AM Jun Wang ***@***.***> wrote: Thank you, I will create an issue for testing fms 2023.03 in UFS WM. — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFCITFCSP5H4F5JAPETYDD3LLAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTOOJVGAZDKNZRGA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-11-22T15:18:05Z

@Hang-Lei-NOAA would you please let us know the location of the following library on wcoss2

fms/2023.02.01
esmf/8.5.0
mapl/2.40.3

Currently we have fms/2023.01, esmf/8.4.2 and mapl/2.35.2 along with other libraries under:

prepend_path("MODULEPATH", "/apps/test/hpc-stack/i-19.1.3.304__m-8.1.12__h-1.14.0__n-4.9.2__p-2.5.10__e-8.4.2/modulefiles/compiler/intel/19.1.3.304")
prepend_path("MODULEPATH", "/apps/test/hpc-stack/i-19.1.3.304__m-8.1.12__h-1.14.0__n-4.9.2__p-2.5.10__e-8.4.2/modulefiles/mpi/intel/19.1.3.304/cray-mpich/8.1.12")

Thanks

climbfuji · 2023-11-22T15:19:43Z

@junwang-noaa we wanted to go to fms@2023.03 to avoid the gnu issues with 2023.02.01? We just installed fms@2023.03 everywhere (see JCSDA/spack-stack#860)

junwang-noaa · 2023-11-22T15:32:35Z

@climbfuji you are correct. My understanding is that right now the fms/2023.02.01(with feature required for GFSv17) is already available in hpc-stack on wcoss2. We did not know the issue with gnu in fms/2023.02.01 on R&D machine when making request to install it on wcoss2. @Hang-Lei-NOAA needs to make a request to install fms@2023.03 on wcoss2, hope that can be installed soon.

climbfuji · 2023-11-22T15:33:32Z

Thanks for the clarification! Makes sense.

Hang-Lei-NOAA · 2023-11-22T15:41:18Z

@jun Wang - NOAA Federal ***@***.***> The fms/2023.02.01 has been in prod. Please check the system installations on wcoss2. Bongi is till working on these esmf/8.5.0 On acorn module load PrgEnv-intel module load intel module load craype module load cray-mpich module ava module load esmf-B/8.5.0 module load mapl-B/2.40.3

…

On Wed, Nov 22, 2023 at 10:33 AM Dom Heinzeller ***@***.***> wrote: Thanks for the clarification! Makes sense. — Reply to this email directly, view it on GitHub <#1874 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AKWSMFFKMYMRQDTSTNP6XDTYFYLNNAVCNFSM6AAAAAA3U5IYJ6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMRSHE4TMNRZG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

junwang-noaa · 2023-12-06T20:18:53Z

@jkbk2004 @FernandoAndrade-NOAA would you please confirm the fms/2023.02.01 is installed on all the R&D platforms? Thanks

climbfuji · 2023-12-06T20:21:10Z

it is in spack-stack-1.5.1, see https://github.com/JCSDA/spack-stack/blob/1.5.1/configs/templates/unified-dev/spack.yaml

climbfuji added the enhancement New feature or request label Aug 18, 2023

climbfuji mentioned this issue Aug 21, 2023

Update to ESMF 8.5.0? #1875

Closed

ulmononian mentioned this issue Aug 23, 2023

Various version updates for spack-stack-1.5.0 (parallelio, fms, met, metplus, eckit, fckit, fiat, ecmwf-atlas, scotch), set boost C++ std to C++-17 JCSDA/spack-stack#722

Merged

5 tasks

junwang-noaa added this to EMC backlog and Model infrastructure FY23Q3-Q4 Aug 28, 2023

junwang-noaa moved this to Todo in Model infrastructure FY23Q3-Q4 Aug 28, 2023

junwang-noaa removed this from EMC backlog Aug 28, 2023

junwang-noaa added this to Model infrastructure development FY24Q1 Sep 28, 2023

junwang-noaa moved this to Todo in Model infrastructure development FY24Q1 Sep 28, 2023

junwang-noaa self-assigned this Oct 23, 2023

junwang-noaa moved this from Todo to In Progress in Model infrastructure development FY24Q1 Oct 23, 2023

climbfuji changed the title ~~Update to fms@2023.02?~~ Update to fms@2023.02.01 Oct 23, 2023

climbfuji mentioned this issue Nov 6, 2023

release/1.5.1: Add fms@2023.03 JCSDA/spack#363

Merged

3 tasks

junwang-noaa mentioned this issue Dec 21, 2023

Spack-stack 1.5.1, ESMF 8.5.0, FMS 2023.02.01 + Remove Gaea C4 + Fix build system to allow CMAKE_<COMPILER>_FLAGS to be specified for submodules #2052 #2013

Merged

43 tasks

jkbk2004 closed this as completed in #2013 Jan 8, 2024

github-project-automation bot moved this from In Progress to Done in Model infrastructure development FY24Q1 Jan 8, 2024

Update to fms@2023.02.01 #1874

Update to fms@2023.02.01 #1874

Comments

climbfuji commented Aug 18, 2023

Description

Solution

Alternatives

Related to

junwang-noaa commented Aug 18, 2023

climbfuji commented Aug 18, 2023 • edited Loading

junwang-noaa commented Aug 28, 2023

junwang-noaa commented Sep 11, 2023

climbfuji commented Sep 11, 2023

junwang-noaa commented Sep 11, 2023

climbfuji commented Sep 11, 2023

Hang-Lei-NOAA commented Sep 11, 2023 via email

Hang-Lei-NOAA commented Oct 4, 2023 via email

junwang-noaa commented Oct 13, 2023

Hang-Lei-NOAA commented Oct 13, 2023 via email

Hang-Lei-NOAA commented Oct 13, 2023 via email

junwang-noaa commented Oct 13, 2023

Hang-Lei-NOAA commented Oct 13, 2023 via email

junwang-noaa commented Oct 13, 2023

junwang-noaa commented Oct 23, 2023

climbfuji commented Oct 23, 2023

Hang-Lei-NOAA commented Oct 24, 2023 via email

junwang-noaa commented Oct 25, 2023

Hang-Lei-NOAA commented Oct 25, 2023 via email

Hang-Lei-NOAA commented Oct 25, 2023 via email

junwang-noaa commented Oct 26, 2023

Hang-Lei-NOAA commented Oct 26, 2023 via email

Hang-Lei-NOAA commented Oct 26, 2023 via email

junwang-noaa commented Oct 26, 2023

junwang-noaa commented Oct 26, 2023

Hang-Lei-NOAA commented Oct 27, 2023 via email

junwang-noaa commented Oct 27, 2023

Hang-Lei-NOAA commented Nov 1, 2023 via email

junwang-noaa commented Nov 2, 2023

junwang-noaa commented Nov 6, 2023

junwang-noaa commented Nov 6, 2023

climbfuji commented Nov 6, 2023

junwang-noaa commented Nov 6, 2023

Hang-Lei-NOAA commented Nov 6, 2023 via email

junwang-noaa commented Nov 22, 2023

climbfuji commented Nov 22, 2023

junwang-noaa commented Nov 22, 2023

climbfuji commented Nov 22, 2023

Hang-Lei-NOAA commented Nov 22, 2023 via email

junwang-noaa commented Dec 6, 2023

climbfuji commented Dec 6, 2023

climbfuji commented Aug 18, 2023 •

edited

Loading